A potential scheme is proposed for generating cluster states of many atoms in cavity quantum electradynamics （QED）, in which an unorthodox encoding is employed with the ground state being qubit [0〉 while two closel...A potential scheme is proposed for generating cluster states of many atoms in cavity quantum electradynamics （QED）, in which an unorthodox encoding is employed with the ground state being qubit [0〉 while two closely spaced upper states being qubit ｜1〉. Throughout the scheme the cavities can be in thermal states but axe only virtually excited. We show how to create the cluster states by performing a two-step hut no single-qubit operation. Discussion is also carried out on the experimental feasibility of our scheme.展开更多
We propose a method to construct an optical cluster-state analyzer based on cross-Kerr nonlinearity combined with linear optics elements. In the scheme, we employ two four-qubit parity gates and the controlled phase g...We propose a method to construct an optical cluster-state analyzer based on cross-Kerr nonlinearity combined with linear optics elements. In the scheme, we employ two four-qubit parity gates and the controlled phase gate （CPG） from only the cross-Kerr nonlinearity and show that all the orthogonal four-qubit cluster states can be completely identified. The scheme is significant for the large-scale quantum communication and quantum information processing networks. In addition, the scheme is feasible and deterministic under current experimental conditions.展开更多
针对高比例新能源并网下光伏集群功率预测面临的复杂空间关联量化建模、功率波动空间传播及技术异质性表征等挑战,该文提出一种基于图状态空间模型(graph state space model,GSSM)的短期同预测方法。首先,通过可学习权重自适应融合光伏...针对高比例新能源并网下光伏集群功率预测面临的复杂空间关联量化建模、功率波动空间传播及技术异质性表征等挑战,该文提出一种基于图状态空间模型(graph state space model,GSSM)的短期同预测方法。首先,通过可学习权重自适应融合光伏电站的地理邻近性、技术相似性与电气耦合性构建多尺度图,精准量化电站间动态关联;其次,建立连续时间状态空间方程,显式引入图扩散项,以物理驱动方式刻画气象扰动在集群中的连续传播过程;最后,利用选择性扫描机制,依据气象特征与电站元数据动态调整模型参数,增强对瞬态气象事件的响应能力,并通过多步预测器完成光伏集群的短期协同预测。基于公开数据集PVOD的实验与算例分析表明,GSSM方法在训练效率、预测精度及稳定性上均显著优于长短时记忆(long short-term memory,LSTM)网络、时空图卷积网络(spatio-temporal graph convolutional networks,STGCN)等基线模型方法,并能有效适应异质性光伏集群场景。该研究可为电网调度提供有力技术支持,助力高比例新能源系统安全运行。展开更多
基金Project supported by the National Natural Science Foundation of China ( Grant Nos 10474118 and 60490280)the Hubei Provincial Foundation for distinguished scholarsthe National Basic Research Program of China (Grant Nos 2005CB724502 and2006CB921203)
文摘A potential scheme is proposed for generating cluster states of many atoms in cavity quantum electradynamics （QED）, in which an unorthodox encoding is employed with the ground state being qubit [0〉 while two closely spaced upper states being qubit ｜1〉. Throughout the scheme the cavities can be in thermal states but axe only virtually excited. We show how to create the cluster states by performing a two-step hut no single-qubit operation. Discussion is also carried out on the experimental feasibility of our scheme.
基金supported by the National Natural Science Foundation of China (Grant Nos. 60667001 and 11165015)
文摘We propose a method to construct an optical cluster-state analyzer based on cross-Kerr nonlinearity combined with linear optics elements. In the scheme, we employ two four-qubit parity gates and the controlled phase gate （CPG） from only the cross-Kerr nonlinearity and show that all the orthogonal four-qubit cluster states can be completely identified. The scheme is significant for the large-scale quantum communication and quantum information processing networks. In addition, the scheme is feasible and deterministic under current experimental conditions.
文摘针对高比例新能源并网下光伏集群功率预测面临的复杂空间关联量化建模、功率波动空间传播及技术异质性表征等挑战,该文提出一种基于图状态空间模型(graph state space model,GSSM)的短期同预测方法。首先,通过可学习权重自适应融合光伏电站的地理邻近性、技术相似性与电气耦合性构建多尺度图,精准量化电站间动态关联;其次,建立连续时间状态空间方程,显式引入图扩散项,以物理驱动方式刻画气象扰动在集群中的连续传播过程;最后,利用选择性扫描机制,依据气象特征与电站元数据动态调整模型参数,增强对瞬态气象事件的响应能力,并通过多步预测器完成光伏集群的短期协同预测。基于公开数据集PVOD的实验与算例分析表明,GSSM方法在训练效率、预测精度及稳定性上均显著优于长短时记忆(long short-term memory,LSTM)网络、时空图卷积网络(spatio-temporal graph convolutional networks,STGCN)等基线模型方法,并能有效适应异质性光伏集群场景。该研究可为电网调度提供有力技术支持,助力高比例新能源系统安全运行。
